CN1663526A - Method for removing Gibbs ring pseudo-image generated in MRI reconstruction process - Google Patents
Method for removing Gibbs ring pseudo-image generated in MRI reconstruction process Download PDFInfo
- Publication number
- CN1663526A CN1663526A CN 200510033773 CN200510033773A CN1663526A CN 1663526 A CN1663526 A CN 1663526A CN 200510033773 CN200510033773 CN 200510033773 CN 200510033773 A CN200510033773 A CN 200510033773A CN 1663526 A CN1663526 A CN 1663526A
- Authority
- CN
- China
- Prior art keywords
- matrix
- data
- image
- frequency domain
- data matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Analysis (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses an eliminating method for Gibbs ring pseudoshadow when reconstructed MRI, and comprises steps as follows: (1) acquiring some data from K space; (2) transforming acquired vector array of original frequency domain; (3) filtrating and getting vector array data 2 of frequency domain from transmitted vector array data 1 of frequency domain; (4) taking edge detection using vector array data 2 of frequency domain; (5) rewriting the image function equation to imaging equation selected second type Chebyshev polynomial as basic functions; (6) by edge information from step (4), substituting vector array data 1 of frequency domain to imaging equation from step (5) in every continuous domain, obtaining needed image domain data array, then transmitting it to vector, resulting in corrected image. The invention can eliminate efficient Gibbs ring pseudoshadow, meantime ensures high resolution for reconstructed image.
Description
Technical field
The present invention relates to the removing method of pseudo-shadow in magnetic resonance (MR) image, especially relate to a kind of removing method that utilizes the Partial K spatial data to carry out magnetic resonance image (MRI) Gibbs ring artifact that produces when rebuilding.
Background technology
Because MR has no radiation injury, resolution advantages of higher, so the application in clinical diagnosis more and more widely.In nuclear magnetic resonance, all use the Fourier formation method usually,, obtain limited initial data because the restriction of imaging time and physics (mainly being gradient intensity) generally can only be carried out the limited number of time phase code.Use Fourier transform to rebuild to limited initial data, common Gibbs ring artifact can occur in reconstructed image, this pseudo-shadow will bring uncertainty for the resolution of organizing details.
Has positive meaning for the reconstruction of adopting part k spatial data to carry out magnetic resonance image (MRI): on the one hand because the minimizing of data acquisition quantity has effectively remedied the defective of nuclear magnetic resonance overlong time; On the other hand, since the time shorten also the possibility of the pseudo-shadow (for example motion artifacts) of other type that is produced in can corresponding minimizing imaging process.So, eliminate the pseudo-shadow that is produced when utilizing the Partial K spatial data to carry out magnetic resonance image (MRI) and have practical meaning.
Two class methods are adopted in elimination for the Gibbs ring artifact usually.The first kind is initial data to be handled the method for general employing filtering.Thereby specific practice is exactly to reduce the Gibbs ring artifact by weakening high-frequency information.Yet because high-frequency data often carries important marginal information, filtering inevitably can cause the ill-defined shortcoming of reconstructed image, and is bigger to the reconstructed image resolution loss like this.Second class is to reach the purpose of eliminating pseudo-shadow by changing algorithm for reconstructing.Lid root fort (Gegenbauer) method for reconstructing is wherein up-to-date, and more a kind of.The Gegenbauer method for reconstructing is in document David and Shu Qiwang " this phenomenon of jeep and its elimination " society's industry and applied mathematics magazine by David (David), propose in 1997 the 39th volume 644～668 pages of fourth phases, after be applied to a kind of in the image and can keep the high-resolution method for reconstructing of reconstructed image.The image border performance that it not only can effectively be eliminated after Gibbs ring artifact and the reconstruction is good.The key of lid root fort (Gegenbauer) method for reconstructing is that wherein parameter is selected, and it is bigger to the reconstructed image quality influence.And it is long to compare traditional Fourier method for reconstructing reconstruction time.In the document IEEE Flame Image Process magazine that " reduces the method for this ring artifact of jeep under a kind of condition that in magnetic resonance imaging, is keeping the organization edge integrity " Volume Four in 2002, author's Rake and Anne efficiently solve the long problem of Gegenbauer method reconstruction time, but the influence of selecting for parameter does not provide solution.
Summary of the invention
The object of the present invention is to provide a kind of can effectively the elimination to utilize the Partial K spatial data to carry out the method for the Gibbs ring artifact that produced in the magnetic resonance image (MRI) process of reconstruction.
Purpose of the present invention can realize by following technical measures, may further comprise the steps successively:
(1) obtains the original frequency domain vectors matrix data of the part that is used for nuclear magnetic resonance, i.e. Partial K spatial data;
(2) be the Kx direction with the frequency coding direction, phase-encoding direction is the Ky direction, is decided to be Ky when phase encoding gradient Gy is zero and equals zero and determine the K space coordinates; The cyclophysis of utilization Fourier transform, with each original frequency domain vectors matrix conversion of obtain become in the K space coordinates with phase-encoding direction Ky equal zero be axis of symmetry grip symmetric matrix data 1 up and down altogether;
(3) frequency domain vectors matrix data 1 is carried out a low-pass filtering and obtain filtered frequency domain vectors matrix data 2;
(4) utilize 2 pairs of images of frequency domain vectors matrix data to carry out rim detection;
(5) be traditional Fourier imaging equation that the image function equation is rewritten into the second class Chebyshev multinomial with fourier progression expanding method be that basic function carries out unfolded image imaging equation;
(6) marginal information of obtaining by step (4) in each continuum, is brought frequency domain vectors matrix data 1 into try to achieve in the step (5) image imaging equation, obtains the image area data asked with this;
(7) will obtain the image area data matrix at last, and vector will be converted into vector, the image that obtains revising;
The concrete steps of frequency domain filtering of the present invention are as follows:
(1) selection index low pass filter in the filtering, have two parameters to need to select in the mathematical model of wave filter: one is filtering strength α, and another is attenuation rate coefficient p; Carry out Filtering Processing under the prerequisite in line with the maintenance edge on parameter is selected, so the attenuation rate coefficient should not selected too small should the selection between 4～8, otherwise image edge information can be destroyed; And factor alpha is generally selection between 25～50 of empirical value;
(2) according to above-mentioned rule structure filtering data matrix, matrix size is consistent with frequency domain vectors matrix data 1;
(3) wait until filtered frequency domain vectors matrix data 2 with multiply each other corresponding of frequency domain vectors data matrix 1 with filtering matrix;
Rim detection is to carry out rim detection respectively to obtain a marginal information on the direction on frequency coding Kx direction and phase code Ky direction among the present invention.The concrete steps that rim detection realizes are as follows:
(1) on frequency coding Kx direction, selected concentration factor τ, the concentration factor data matrix of formation.
(2) utilize concentration factor structure to concentrate kernel function K, make and concentrate kernel function K to have following characteristic: carry out the jump that its value behind the convolution algorithm converges on function to be detected with image function;
(3) to concentrating the nuclear matrix data to carry out Fourier transform, with data matrix after the conversion and frequency domain data matrix 2 corresponding multiplying each other, the data matrix that obtains carries out inverse fourier transform again;
(4) setting threshold is got rid of the false discontinuous point in the detection, has only test value just to be judged to be real testing result greater than the result of threshold value; The size of threshold value is 9% of the maximum jump value of function;
(5) matrix data that carries out obtaining after the threshold test is the final result of frequency coding Kx direction edge detection;
(6) according to above-mentioned steps phase code Ky direction is detected matrix data after obtaining detecting;
The concrete steps that the construct image imaging equation is realized among the present invention are as follows:
(1) respectively in frequency coding Kx direction, the selected second class Chebyshev multinomial U (x) on the phase code Ky direction, U (y);
(2) data matrix that utilizes Chebyshev multinomial U (x) to constitute carries out Fourier and changes and obtain transformation matrix w (x), and data matrix U (y) carries out Fourier transform and obtains transformation matrix w (y);
(3) on frequency coding direction and phase-encoding direction, the transform data matrix is done inverse transformation and obtain w
-1(x), w
-1(y);
(4) the inverse transformation data matrix w that obtains
-1(x), w
-1(y) obtain image imaging equation expansion coefficient matrix g with frequency domain vectors data matrix 1 corresponding multiplying each other respectively
l(x), g
l(y);
(5), use function expansion coefficient data matrix g on the phase code Ky direction respectively in frequency coding Kx direction
l(x), g
l(y) with Chebyshev polynomial data matrix U (x), U (y) correspondence multiplies each other and can obtain image imaging equation G on the frequency coding direction
l(x) and the image imaging equation G on the phase-encoding direction
l(y).
The present invention's's " based on polynomial reverse method for reconstructing of Chebyshev " advantage is as follows:
1. eliminating the high-resolution that the Gibbs ring artifact has guaranteed reconstructed image simultaneously;
2. compare the method for the initial data filtering of elimination Gibbs ring artifact commonly used, reconstructed image spatial resolution of the present invention is higher, edge clear;
3. compare existing Gegenbauer method for reconstructing, it has effectively shortened reconstruction time under the prerequisite that effectively improves the reconstruction precision;
4. in process of reconstruction, need not artificially select parameter;
Description of drawings:
Fig. 1 flow chart of the present invention;
Fig. 2 .1 template original image,
The pseudo-image that Fig. 2 .2 reconstructs with part k spatial data for simulation,
Fig. 2 .3 eliminates the image that reconstructs behind the Gibbs ring artifact for utilization the inventive method
Fig. 3 .1 original image,
Fig. 3 .2 is the image after the pseudo-shadow of simulation is rebuild,
Fig. 3 .3 eliminates pseudo-movie queen's image for using the present invention.
A direction edge detection of Fig. 4 algorithm flow chart;
Fig. 5 produces the imaging equation algorithm flow chart.
The specific embodiment:
The present invention is described in further detail below in conjunction with accompanying drawing,
The original frequency domain vectors matrix data of the part of magnetic resonance image (MRI) can be from magnetic resonance imaging system directly obtains, and also the primary template image can be carried out Fourier transform and simulate and obtain.As Fig. 2, size simulates the required frequency domain vectors matrix data of imaging for the primary template image of 256*256 carries out Fourier transform to it, the frequency domain vectors matrix data is carried out symmetry respectively block 60 row high-frequency datas, and the data of disappearance are filled up with 0.Carry out the inverse fourier transform reconstructed image with part frequency domain vectors data then and simulate the image that contains the Gibbs ring artifact.To the correction of pseudo-image successively as follows:
1. be the Kx direction with the frequency coding direction, phase-encoding direction is the Ky direction, is decided to be Ky when phase encoding gradient Gy is zero and equals zero and determine the K space coordinates; The cyclophysis of utilization Fourier transform, with each original frequency domain vectors matrix conversion of obtain become in the K space coordinates with phase-encoding direction Ky equal zero be axis of symmetry grip symmetric matrix data 1 up and down altogether
2. the selection index low-pass filtering is constructed filtering matrix, and selecting the value of filtering strength α is 40, and attenuation rate coefficient p is 5.
3. wait until filtered frequency domain vectors matrix data 2 with multiply each other corresponding of frequency domain vectors data matrix 1 with filtering matrix;
4. filtered frequency domain vectors matrix 2 is carried out rim detection with regard to Frequency Domain Coding Kx direction and phase code Ky direction respectively.The specific implementation step is: at first frequency domain vectors matrix after the filtering 2 is carried out the two-dimension fourier inverse transformation, obtain the spatial domain matrix data.The one dimensional fourier transform that the spatial domain matrix data is carried out on the x direction obtains matrix data 3 then.At first selected concentration factor τ, concrete choice criteria is two pieces of documents of face as follows: Anne and Ai Tan " rim detection of spectrum data " use and calculate the 101st～135 page of 1999 the 7th volume of magazine and Anne and Ai Tan " the rim detection II of spectrum data: non-linear enhancing " society's industry and the 1389th～1408 page of applied mathematics magazine calendar year 2001 the 38th volume.In above-mentioned two pieces of articles, the author to the choice criteria of concentration factor and wherein the value of parameter carried out discussing and provided a large amount of experiments and proved.In the present invention, directly employing wherein detects the best index concentration factor of effect, the concentration factor data matrix of formation.Multiply each other corresponding with the data of the every row of concentration factor matrix of matrix data 3 obtained frequency domain data matrix 4, and the inverse fourier transform that data matrix 4 is carried out on the x direction obtains spatial domain data matrix 5 again.Calculating data matrix 5 every row elements peaked 9% are formed the threshold value array successively.Every row element in the data matrix 5 and threshold value array respective value are relatively judged, the element less than threshold value is rewritten as 0 keeps, obtain the spatial domain matrix data and be edge detection results on the final Kx direction greater than the element initial value of threshold value.According to the testing result that obtains with quadrat method on the Ky direction.Idiographic flow such as Fig. 4.
5. according to the marginal information that arrives of above-mentioned steps 4, in continuum, image is rebuild.The concrete steps that the construct image imaging equation is realized are as follows:
(1) respectively in frequency coding Kx direction, the selected second class Chebyshev multinomial U (x) on the phase code Ky direction, U (y);
(2) data matrix that utilizes Chebyshev multinomial U (x) to constitute carries out Fourier and changes and obtain transformation matrix w (x), and data matrix U (y) carries out Fourier transform and obtains transformation matrix w (y);
(3) on frequency coding direction and phase-encoding direction, the transform data matrix is done inverse transformation and obtain w
-1(x), w
-1(y);
(4) the inverse transformation data matrix w that obtains
-1(x), w
-1(y) obtain image imaging equation expansion coefficient matrix g with frequency domain vectors data matrix 1 corresponding multiplying each other respectively
l(x), g
l(y);
(5), use function expansion coefficient data matrix g on the phase code Ky direction respectively in frequency coding Kx direction
l(x), g
l(y) with Chebyshev polynomial data matrix U (x), U (y) correspondence multiplies each other and can obtain image imaging equation G on the frequency coding direction
l(x) and the image imaging equation G on the phase-encoding direction
l(y); Idiographic flow such as Fig. 5;
6. respectively in frequency coding Kx direction, on the phase code Ky direction, by the marginal information of obtaining, promptly utilize the edge detection results data matrix on the resulting both direction in the step 4, in the continuum, utilize frequency domain matrix data 1, calling graph is as imaging equation G
l(x), G
l(y), obtain the image area data matrix F (x) on frequency coding direction and the phase-encoding direction, F (y), with the image area data matrix F (x) on the both direction, F (y) does weighted average and obtains final image numeric field data matrix F; And vector is converted into vector, the image that obtains rebuilding.
Claims (5)
- The removing method of the Gibbs ring artifact that produces when 1, a kind of magnetic resonance image (MRI) is rebuild is characterized in that may further comprise the steps:(1) obtains the original frequency domain vectors matrix data of the part that is used for nuclear magnetic resonance, i.e. Partial K spatial data;(2) be the Kx direction with the frequency coding direction, phase-encoding direction is the Ky direction, is decided to be Ky when phase encoding gradient Gy is zero and equals zero and determine the K space coordinates; The cyclophysis of utilization Fourier transform, with each original frequency domain vectors matrix conversion of obtain become in the K space coordinates with phase-encoding direction Ky equal zero be axis of symmetry grip symmetric matrix data 1 up and down altogether;(3) frequency domain vectors matrix data 1 is carried out a low-pass filtering and obtain filtered frequency domain vectors matrix data 2;(4) utilize 2 pairs of images of frequency domain vectors matrix data to carry out rim detection;(5) be traditional Fourier imaging equation that the image function equation is rewritten into the second class Chebyshev multinomial with fourier progression expanding method be that basic function carries out unfolded image imaging equation;(6) marginal information of obtaining by step (4) in each continuum, is brought frequency domain vectors matrix data 1 into try to achieve in the step (5) image imaging equation, obtains the image area data matrix asked with this;(7) will obtain the image area data matrix at last, and vector will be converted into vector, the image that obtains revising.
- The removing method of the Gibbs ring artifact that produces when 2, magnetic resonance image (MRI) according to claim 1 is rebuild, it is characterized in that: the concrete steps of described frequency domain filtering are as follows:(1) selection index low pass filter in the filtering, wherein the attenuation rate coefficient is selected between 4～8, and the filtering strength factor alpha is that empirical value is generally selected between 25～50;(2) according to above-mentioned rule structure filtering data matrix, matrix size is consistent with frequency domain vectors matrix data 1;(3) wait until filtered frequency domain vectors matrix data 2 with multiply each other corresponding of frequency domain vectors data matrix 1 with filtering matrix.
- The removing method of the Gibbs ring artifact that produces when 3, magnetic resonance image (MRI) according to claim 1 is rebuild, it is characterized in that: described rim detection is to carry out rim detection respectively to obtain a marginal information on the direction on frequency coding Kx direction and phase code Ky direction, and concrete steps are as follows:(1) on frequency coding Kx direction, selected concentration factor τ, the concentration factor data matrix of formation;(2) utilize concentration factor structure to concentrate kernel function K, make and concentrate kernel function K to have following characteristic: carry out the jump that its value behind the convolution algorithm converges on function to be detected with image function;(3) to concentrating the nuclear matrix data to carry out Fourier transform, with data matrix after the conversion and frequency domain data matrix 2 corresponding multiplying each other, the data matrix that obtains carries out inverse fourier transform again;(4) setting threshold is got rid of the false discontinuous point in the detection, has only test value just to be judged to be real testing result greater than the result of threshold value; The size of threshold value is 9% of the maximum jump value of function;(5) matrix data that carries out obtaining after the threshold test is the final result of frequency coding Kx direction edge detection;(6) according to above-mentioned steps phase code Ky direction is detected matrix data after obtaining detecting;
- The removing method of the Gibbs ring artifact that produces when 4, magnetic resonance image (MRI) according to claim 1 is rebuild, it is characterized in that: the concrete steps of construct image imaging equation are as follows:(1) respectively in frequency coding Kx direction, the selected second class Chebyshev multinomial U (x) on the phase code Ky direction, U (y);(2) data matrix that utilizes Chebyshev multinomial U (x) to constitute carries out Fourier and changes and obtain transformation matrix w (x), and data matrix U (y) carries out Fourier transform and obtains transformation matrix w (y);(3) on frequency coding direction and phase-encoding direction, the transform data matrix is done inverse transformation and obtain w -1(x), w -1(y);(4) the inverse transformation data matrix w that obtains -1(x), w -1(y) obtain image imaging equation expansion coefficient matrix g with frequency domain vectors data matrix 1 corresponding multiplying each other respectively l(x), g l(y);(5), use function expansion coefficient data matrix g on the phase code Ky direction respectively in frequency coding Kx direction l(x), g l(y) with Chebyshev polynomial data matrix U (x), U (y) correspondence multiplies each other and can obtain image imaging equation G on the frequency coding direction l(x) and the image imaging equation G on the phase-encoding direction l(y).
- The removing method of the Gibbs ring artifact that produces when 5, magnetic resonance image (MRI) according to claim 1 is rebuild, it is characterized in that: the process that obtains the image area data matrix is: respectively in frequency coding Kx direction, on the phase code Ky direction, by the marginal information of obtaining, calling graph is as imaging equation in the continuum, obtain the image area data matrix F (x) on frequency coding direction and the phase-encoding direction, F (y); With the image area data matrix F (x) on the both direction, F (y) does weighted average and obtains final image numeric field data matrix F.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100337730A CN1327810C (en) | 2005-03-28 | 2005-03-28 | Method for removing Gibbs ring pseudo-image generated in MRI reconstruction process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100337730A CN1327810C (en) | 2005-03-28 | 2005-03-28 | Method for removing Gibbs ring pseudo-image generated in MRI reconstruction process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1663526A true CN1663526A (en) | 2005-09-07 |
| CN1327810C CN1327810C (en) | 2007-07-25 |
Family
ID=35034965
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100337730A Expired - Fee Related CN1327810C (en) | 2005-03-28 | 2005-03-28 | Method for removing Gibbs ring pseudo-image generated in MRI reconstruction process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1327810C (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100409807C (en) * | 2005-11-29 | 2008-08-13 | 东南大学 | Artifact correction method based on even marker in plane echo imaging technique |
| CN101908204A (en) * | 2010-05-25 | 2010-12-08 | 南方医科大学 | Inverse diffusion method for eliminating Gibbs annular artifact of magnetic resonance image |
| CN101995561A (en) * | 2010-10-22 | 2011-03-30 | 南方医科大学 | Reconstruction method of periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) magnetic resonance data based on image domain overlapping |
| CN102096055A (en) * | 2010-12-14 | 2011-06-15 | 南方医科大学 | Rapid and accurate reconstructing method for non-uniform sampling data of magnetic resonance imaging |
| CN102324089A (en) * | 2011-07-13 | 2012-01-18 | 南方医科大学 | Maximum posteriori reconstruction method of PET (positron emission tomography) image based on generalized entropy and MR (magnetic resonance) prior |
| CN103942763A (en) * | 2014-05-03 | 2014-07-23 | 南方医科大学 | Voxel level PET (positron emission tomography) image partial volume correction method based on MR (magnetic resonance) information guide |
| CN104020430A (en) * | 2014-04-15 | 2014-09-03 | 清华大学 | Correction method and system for movement artifacts of magnetic resonance imaging |
| CN104635187A (en) * | 2015-02-28 | 2015-05-20 | 深圳先进技术研究院 | Adaptive active automatic shimming method and system of magnetic resonance system |
| CN110133556A (en) * | 2019-05-29 | 2019-08-16 | 上海联影医疗科技有限公司 | A kind of magnetic resonance image processing method, device, equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IL86570A (en) * | 1988-05-31 | 1991-07-18 | Elscint Ltd | Reduction of truncation caused artifacts |
| US5285157A (en) * | 1991-05-03 | 1994-02-08 | Elscint Ltd. | Reduction of truncation caused artifacts |
| CN1050427C (en) * | 1995-06-17 | 2000-03-15 | 浙江大学 | Fuzzy mutticriteria image reconstruction method |
-
2005
- 2005-03-28 CN CNB2005100337730A patent/CN1327810C/en not_active Expired - Fee Related
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100409807C (en) * | 2005-11-29 | 2008-08-13 | 东南大学 | Artifact correction method based on even marker in plane echo imaging technique |
| CN101908204B (en) * | 2010-05-25 | 2012-01-04 | 南方医科大学 | Inverse diffusion method for eliminating Gibbs annular artifact of magnetic resonance image |
| CN101908204A (en) * | 2010-05-25 | 2010-12-08 | 南方医科大学 | Inverse diffusion method for eliminating Gibbs annular artifact of magnetic resonance image |
| CN101995561A (en) * | 2010-10-22 | 2011-03-30 | 南方医科大学 | Reconstruction method of periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) magnetic resonance data based on image domain overlapping |
| CN102096055B (en) * | 2010-12-14 | 2013-01-23 | 南方医科大学 | Rapid and accurate reconstructing method for non-uniform sampling data of magnetic resonance imaging |
| CN102096055A (en) * | 2010-12-14 | 2011-06-15 | 南方医科大学 | Rapid and accurate reconstructing method for non-uniform sampling data of magnetic resonance imaging |
| CN102324089A (en) * | 2011-07-13 | 2012-01-18 | 南方医科大学 | Maximum posteriori reconstruction method of PET (positron emission tomography) image based on generalized entropy and MR (magnetic resonance) prior |
| CN104020430A (en) * | 2014-04-15 | 2014-09-03 | 清华大学 | Correction method and system for movement artifacts of magnetic resonance imaging |
| CN104020430B (en) * | 2014-04-15 | 2017-01-25 | 清华大学 | Correction method and system for movement artifacts of magnetic resonance imaging |
| CN103942763A (en) * | 2014-05-03 | 2014-07-23 | 南方医科大学 | Voxel level PET (positron emission tomography) image partial volume correction method based on MR (magnetic resonance) information guide |
| CN104635187A (en) * | 2015-02-28 | 2015-05-20 | 深圳先进技术研究院 | Adaptive active automatic shimming method and system of magnetic resonance system |
| CN104635187B (en) * | 2015-02-28 | 2017-07-07 | 深圳先进技术研究院 | The adaptive active automatic shimming method and system of magnetic resonance system |
| CN110133556A (en) * | 2019-05-29 | 2019-08-16 | 上海联影医疗科技有限公司 | A kind of magnetic resonance image processing method, device, equipment and storage medium |
| CN110133556B (en) * | 2019-05-29 | 2021-01-19 | 上海联影医疗科技股份有限公司 | Magnetic resonance image processing method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1327810C (en) | 2007-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1663526A (en) | Method for removing Gibbs ring pseudo-image generated in MRI reconstruction process | |
| CN110490832A (en) | A kind of MR image reconstruction method based on regularization depth image transcendental method | |
| Cheon et al. | Generative adversarial network-based image super-resolution using perceptual content losses | |
| CN105631807B (en) | The single-frame image super-resolution reconstruction method chosen based on sparse domain | |
| CN111932460B (en) | MR image super-resolution reconstruction method, device, computer equipment and storage medium | |
| CN107576925B (en) | Magnetic resonance multi-contrast image reconstruction method and device | |
| CN112396672B (en) | Sparse angle cone-beam CT image reconstruction method based on deep learning | |
| CN107274462A (en) | The many dictionary learning MR image reconstruction methods of classification based on entropy and geometric direction | |
| CN106485764A (en) | The quick exact reconstruction methods of MRI image | |
| CN105118053B (en) | A kind of full reference picture assessment method for encoding quality based on compressed sensing | |
| CN104700440B (en) | Magnetic resonant part K spatial image reconstruction method | |
| CN113160380B (en) | Three-dimensional magnetic resonance image super-resolution reconstruction method, electronic equipment and storage medium | |
| Wang et al. | High‐field mr diffusion‐weighted image denoising using a joint denoising convolutional neural network | |
| US8634671B2 (en) | Methods and apparatus to perform multi-focal plane image acquisition and compression | |
| Brun et al. | Effective implementation of ring artifacts removal filters for synchrotron radiation microtomographic images | |
| CN110942496A (en) | Propeller sampling and neural network-based magnetic resonance image reconstruction method and system | |
| CN117523095A (en) | Sparse angle THz-CT image reconstruction method based on deep learning | |
| CN112150360A (en) | IVUS image super-resolution reconstruction method based on dense residual error network | |
| Schrank et al. | Deep learning MPI super-resolution by implicit representation of the system matrix | |
| CN114897691A (en) | Image compressed sensing reconstruction method and system based on standardized stream | |
| CN115239836A (en) | Extreme sparse view angle CT reconstruction method based on end-to-end neural network | |
| Abbas et al. | PET image reconstruction and denoising on hexagonal lattices | |
| CN116934618B (en) | Image halftone method, system and medium based on improved residual error network | |
| CN112801886B (en) | Dynamic PET image denoising method and system based on image wavelet transformation | |
| CN117036533B (en) | Sparse angle CT reconstruction method and system for coaxial phase contrast imaging |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Assignee: Ningbo Xingaoyi Magnetism Co., Ltd. Assignor: Southern Medical University Contract fulfillment period: 2007.11.15 to 2013.11.14 contract change Contract record no.: 2008330000178 Denomination of invention: Method for removing Gibbs ring pseudo-image generated in MRI reconstruction process Granted publication date: 20070725 License type: Exclusive license Record date: 2008.8.15 |
|
| LIC | Patent licence contract for exploitation submitted for record |
Free format text: EXCLUSIVE LICENCE; TIME LIMIT OF IMPLEMENTING CONTACT: 2007.11.15 TO 2013.11.14 Name of requester: NINGBO XIN GAO YI MAGNETIC CO., LTD. Effective date: 20080815 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20070725 Termination date: 20150328 |
|
| EXPY | Termination of patent right or utility model |